方法证据记录
Self-Attention
Multi-head self-attention, introduced by Vaswani and colleagues in 2017, is the mechanism that lets every position in a sequence compute its relationship to all other positions in parallel. It is the core of the Transformer architecture and the foundation underneath BERT, GPT, and T5.
源记录
引文逐字复制自方法源记录。这些引文不代表任何层级的验证。
Multi-Head Self-Attention (Transformer Core)
分类方法记录 · ml-model / deep-learning
- Vaswani, A. et al. (2017). Attention Is All You Need. NeurIPS. · URL
- Devlin, J. et al. (2019). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. NAACL. · URL
精选声明
声明已持久化到证据分类账中，每个声明都有自己的评估。
尚无精选声明
当分类账中没有声明时，此视图不会自行创建声明评估。
相关方法
从方法图中生成，显示为机器建议的关系 — 不推断任何证据声明。